Add RunEvaluator

[vowelparrot]

See: langchain-ai/langchain#5618 for example LangChain implementations

I think we'll want some simple non-langchain completion function evaluators if we want this core interface to be of much use outside the OSS project. I also don't think I want to land the StringEvaluator class but put it up as an example of one approach (that contrasts with the ones above)

from typing import Tuple, Optional
from langchain.chat_models import ChatOpenAI
from langchain.evaluation.qa import QAEvalChain
from langchainplus_sdk.client import LangChainPlusClient
from langchainplus_sdk.evaluation.evaluator import StringEvaluator

client = LangChainPlusClient()

dataset_name = "calculator-example-dataset-2"
examples = client.list_examples(dataset_name=dataset_name)
runs = [run for example in examples for run in client.list_runs(reference_example=example.id, error=False)]


chain = QAEvalChain.from_llm(llm=ChatOpenAI())
def evaluate(input_: str, output_: str, answer: Optional[str]) -> Tuple[str, Optional[float]]:
    result = chain({"query": input_, "result": output_, "answer": answer})
    return result["text"], None

evaluator = StringEvaluator(
    evaluation_name="Correctness",
    input_key="input",
    prediction_key='output',
    answer_key='output', 
    grading_function=evaluate
)

for run in runs:
    client.evaluate_run(run, evaluator)

[agola11]

looks good!